Hip fracture is a sentinel acute event in the life of an older person. While not classically studied as an end-of- life condition, it frequently precipitate a transition to functional dependence and death. Current paradigms of hip fracture research and clinical care have been informed by a paradigm in which the primary goal of care is to return patients to their pre-fracture level of functioning. This is a very worthy goal, but it needs to be balanced by the understanding that the majority of persons with hip fracture do not return to their prior level of functioning. We need treatment paradigms that view supportive and palliative treatment of patients with poor outcomes following hip fracture as important goals of care. However, the ability to provide supportive and palliative services to elders following hip fracture is hampered because there is virtually no population-based data about what those needs are who is most likely to have poor functional recovery following hip fracture, and the types of care currently received by patients who do have poor health outcomes following hip fracture. To address this gap, we will create the information database that is needed to guide policy makers and those planning intervention studies that will better address the supportive and palliative needs of hip fracture patients who do not return to their prior level of functioning. Specifically,we will create a nationally representative database of older persons with hip fracture to address three primary questions. First, how does the burden of hip fracture affect the need for personal assistance with basic and instrumental activities of daily living? Second, what are the predictors of poor outcomes following hip fracture? Third, what is the current end-of- life care patterns for elders who die in the year after hip fracture? We propose opportunistic use of the Health and Retirement Study (HRS) to better define supportive and palliative needs of elders with hip fracture. Several compelling features of the HRS make it ideal for this study. First, because of it size and the considerable follow-up, the HRS makes it possible to study large numbers of elders with hip fracture with extensive data on their health, functional status, and care needs both before and after hip fracture. Second, the HRS has extensive data on medical and psychosocial domains of risk that will enhance the ability to identify those at highest risk of poor outcomes. Third, because the HRS is linked to Medicare, we can paint a detailed description of the type of care received towards the end of life. Fourth, because it interviews next of kin after death, we will be able to learn about the quality of symptom control and elements of end-of-life planning. The HRS allows us to conduct a study that would be impossible using primary data collection approaches in a cost-effective manner. We propose a cohort study with the following aims: Aim 1: Describe the need for personal assistance in basic and instrumental activities of daily living i the two years before and the two years after hip fracture. Aim 2: Determine which elders are at highest risk for death and disability following hip fracture. Aim 3: Among those who die in the year after hip fracture, describe patterns of resource use (hospital, ICU, and hospice), location of death, advance care planning, and quality of symptom control.